Drainage density is a quantitative measure used to assess the spacing of channels within stream networks (Tamiru and Wagari, 2022). It provides valuable insights into the proximity and distribution of these channels within a given geographical area. Assessing drainage density is crucial for understanding groundwater potential, as there is a well-established inverse relationship between drainage density and permeability (Tamesgen et al., 2023). In this study, data analysis was conducted utilizing a Digital Elevation Model (DEM) within the ArcGIS 10.8 framework, employing the line density interpolation tool to derive results. The investigation yielded a spectrum of drainage density values ranging from 0 to 465.5 km2. These values were subsequently reclassified into five subclasses, as illustrated in Fig. 1. The subclasses include very low density (0-60.3 km2), low density (60.3-118.7 km2), moderate density (118.8-178.9 km2), high density (179.0-255.6 km2), and very high density (255.7-465.5 km2). Areas characterized by very low drainage density are indicative of a high infiltration ratio, suggesting favourable conditions for groundwater potential. Consequently, such areas carry the highest weighting in the analysis. Conversely, regions exhibiting exceptionally high drainage density were assigned the lowest weighting due to their diminished potential for groundwater resources. These findings underscore the significance of drainage density assessments in delineating areas with varying groundwater potential, thereby informing effective resource management strategies.
Stream order, as determined by the Strahler stream ordering technique, classifies streams within a watershed based on their hierarchy and connectivity (Shekar and Mathew, 2023). Streams of the first order have no tributaries, while higher-order streams result from the confluence of lower-order streams (Erosemiah and Viji, 2023). The importance of stream order for groundwater potential zonation lies in its reflection of the drainage network's complexity and connectivity. Higher stream orders (3, 4, and 5) typically indicate a more developed and interconnected drainage system, suggesting increased potential for groundwater recharge due to enhanced surface water interaction with subsurface aquifers (Shekar and Mathew, 2023). Conversely, lower stream orders (1 and 2) denote areas with limited groundwater recharge potential, as they represent smaller, less developed drainage systems with reduced interaction with aquifers. Understanding the distribution and characteristics of different stream orders facilitates the delineation of GWPZs, enabling the identification of areas where groundwater resources are more abundant and accessible, thus informing effective water resource management strategies.
Lineaments, which encompass faults, fractures, and joints on the Earth's surface, serve as conduits for groundwater movement and offer insights into subsurface faults and fractures (Islam et al., 2023). These linear structures play a crucial role in identifying secondary structural abnormalities, with data from digital remote sensing aiding in their identification. Areas surrounding lineaments and their intersections are considered ideal for groundwater storage due to their high infiltration capacity (Zewdie et al., 2024). As one moves away from lineaments, the intensity of fractures diminishes, indicating that lineaments are prime targets for groundwater (Zewdie et al., 2024). They are pivotal in identifying potential groundwater reservoirs, serving as pathways for movement and storage within subsurface aquifers (Faheem et al.,2023). In this study, lineament density was determined using a Digital Elevation Model (DEM) in conjunction with the lineament density tool in ArcGIS 10.8. These values were then categorized into five distinct subclasses, as illustrated in Fig. 2: very low (0-0.18 km/km2), low (0.19–0.49 km/km2), medium (0.50–0.80 km/km2), high (0.81–1.16 km/km2), and very high (1.17–1.93 km/km2). This study's findings closely align with those of Shelar et al. (2023), who investigated the Urmodi River basin in Maharashtra, India's Western Ghats. They used eight parameters to delineate GWPZs, aiming to gather and compile thematic maps from satellite and field data, assess and assign weights to each thematic layer, and generate a GWPZs map using GIS and AHP methodologies. The delineation of these subclasses allows for a comprehensive understanding of the distribution and intensity of lineament features within the study area. Areas characterized by higher lineament density values, such as those classified as high or very high, are indicative of increased potential for groundwater resources due to the enhanced connectivity and permeability provided by the presence of numerous lineaments (Ozegin et al., 2023). Conversely, areas with lower lineament density values exhibit reduced groundwater potential, as the scarcity of linear features impedes groundwater flow and storage within the subsurface geological formations (Ozegin et al., 2023).
The study area encompasses two distinct soil types: eutric fluvisols and cryic soils. Eutric fluvisols are characterized by their fertility and high nutrient content, typically found in floodplains and river valleys where they are deposited by flowing water (Rodrigo-Comino et al., 2023). These soils are well-drained and exhibit good moisture retention properties, making them conducive to agriculture and vegetation growth (Rodrigo‐Comino et al., 2023). Their presence often signifies areas with abundant groundwater potential, as the fertile soil supports robust plant growth, which in turn indicates the presence of water sources (Furtak et al., 2024). Cryic soils, typically associated with cold climates featuring permafrost, are also present in hot regions. Despite containing inherent moisture, these soils have limited water availability in warmer areas due to permafrost melting (Raudina et al.,2023). While they are crucial for identifying GWPZs, their significance is diminished in hot regions where water availability is constrained (Raudina et al.,2023). Eutric fluvisols, fertile and well-drained, indicate accessible groundwater suitable for agriculture (Furtak et al., 2024). Cryic soils contribute to groundwater recharge in cold regions but have limited significance in hot and dry areas due to climate conditions (Raudina et al.,2023).
A slope denotes the incline or gradient of a surface, measured by the ratio of vertical change (rise) to horizontal distance (run) (Kadlíček and Mašín, 2023). It characterizes the steepness of the terrain and can be expressed as a percentage, angle, or ratio (Kadlíček and Mašín, 2023). Slope exerts a pivotal influence on natural processes like erosion, runoff, and soil stability, profoundly impacting the suitability of land for agricultural, construction, and transportation purposes (Lann et al., 2024). Within the study area, slopes play a pivotal role in determining GWPZs s. With increasing slope angle, surface runoff rises, diminishing infiltration and recharge capacity. This inverse relationship arises from the swifter surface flow on steeper slopes, impeding water retention and soil infiltration. The slope map, generated through terrain analysis using DEM within the ArcGIS 10.8 environment, reveals slopes ranging from 00 to 1.070 (Fig. 4). To further elucidate the variations in slope characteristics, the values were reclassified into five distinct subclasses: flat slope areas (00-1.070), gentle slope areas (1.080–1.890), medium slope areas (1.90–2.810), steep slope areas (2.820–4.260), and very steep slope areas (4.270–12.360). Areas characterized by flat and gentle slope areas exhibit higher infiltration ratios due to their reduced surface runoff and longer residence time for water to infiltrate and recharge the saturated zones (Pocco et al., 2023). Consequently, these areas were assigned the highest weighting in the analysis. Conversely, regions with very steep slopes experience rapid surface runoff, limiting infiltration and recharge capabilities (Pocco et al., 2023). Hence, these areas were assigned the lowest weighting due to their diminished groundwater potential.
The precipitation in Botswana reveals spatial variability, with annual rainfall ranging from a minimum of 275.0 mm to 1161.31 mm (Kwembeya and Shikangalah, 2023). The mean annual rainfall varies from over 650 mm in the northeast to less than 250 mm in the southwest, with specific regional ranges such as 620 mm in the northern Kasane area to 300 mm in the southwestern Tsabong area, averaging 475 mm annually nationwide (Chikuta et al., 2024). However, recent data highlights variations within the country. In 2021, southwestern Botswana experienced average minimum rainfall ranging from about 275 mm to 456 mm in the southwestern region (Fig. 5). This was followed by 456 mm to 605 mm in the central southern and eastern areas. The south-eastern, southwestern, and north-western regions received 605 mm to 734 mm on average in 2021. Within the study areas, rainfall is relatively uniform, ranging from 734 mm to 911 mm annually. Notably, the highest rainfall occurred in northern Botswana, where 911 mm to 1161 mm was recorded in 2021. Understanding the impact of rainfall variations on groundwater potential is crucial, as it involves recognizing the intricate relationship between precipitation patterns and groundwater recharge rates (Shekar and Mathew, 2023). Higher rainfall regions facilitate greater infiltration of water into the soil, leading to increased aquifer recharge and subsequently higher groundwater potential (Kodihal and Akhtar, 2024). Conversely, lower rainfall areas may experience limited groundwater recharge rates, resulting in diminished groundwater potential. Moreover, the distribution of rainfall events influences groundwater recharge differently. Intense, short-duration rainfall events lead to surface runoff, reducing the amount of water available for infiltration and recharge. (Bresinsky et al., 2023). In contrast, steady, prolonged rainfall allows for deeper penetration into the soil, promoting groundwater recharge and enhancing groundwater potential (Xu et al., 2024). Influence on GWPZs extends beyond rainfall variations; land use and land cover play significant roles as well (Shekar and Mathew, 2023). These factors interact intricately, shaping groundwater dynamics. For instance, land use practices, such as agricultural irrigation or urban development, can alter the natural hydrological cycle, affecting groundwater recharge rates. Moreover, geological factors, such as soil composition and bedrock permeability, further modulate the impact of rainfall on groundwater potential. Therefore, understanding the interplay between land use, geological characteristics, and rainfall patterns is essential for accurately assessing and managing groundwater resources.
Urbanization, deforestation, and unsustainable agricultural practices increase surface runoff and reduce infiltration rates, negatively impacting groundwater recharge (Vijay and Varija, 2024). Conversely, areas with permeable geological formations, such as sand and gravel aquifers, are more conducive to groundwater recharge, resulting in higher groundwater potential even with lower rainfall amounts (Vijay and Varija, 2024; Shekar and Mathew, 2023). Thus, variations in rainfall patterns across Botswana have significant implications for groundwater potential. Understanding these relationships is crucial for effective water resource management and sustainable utilization of groundwater in different parts of the country.
The LULC significantly influences groundwater dynamics through alterations in recharge and demand. Additionally, groundwater interacts intricately with landscape features and land use practices. The quantity, timing, and distribution of groundwater recharge are modulated by land cover types, which directly affect factors such as runoff and evapotranspiration. To assess these relationships, LULC layers were generated utilizing Landsat 8 OLI/TIRS imagery, and a Random Tree Classifier was employed within the ArcGIS 10.8 software platform. The study area encompasses six primary LULC classes. Grasslands dominate the study area, constituting 39.3% (289 km2) of its total extent. Following closely are areas characterized by healthy shrubs and trees, covering 24.5% (180 km2) of the study area. Barren land, gravel roads, and tarred roads collectively occupy 32.7% (240.8 km2) of the study area. Built-up areas represent a minor portion, covering 2.3% (16.8 km2) of the study area. Finally, water bodies occupy the smallest proportion, accounting for only 1.2% of the study area. Each LULC component exerts a distinct influence on GWPZs. Grasslands, with their relatively permeable surfaces, facilitate infiltration and groundwater recharge, making them favourable for groundwater replenishment. Healthy shrubs and trees similarly contribute to groundwater recharge by intercepting rainfall, reducing runoff, and promoting infiltration. Barren land, gravel roads, and tarred roads impede groundwater recharge due to their impermeable nature, leading to increased surface runoff and reduced infiltration rates. Built-up areas often feature extensive impervious surfaces such as roads and buildings, further limiting groundwater recharge. Conversely, water bodies serve as direct contributors to groundwater recharge, particularly in areas with high groundwater-table levels. Understanding the distribution and characteristics of these LULC components is essential for delineating GWPZs and formulating effective groundwater management strategies. By considering the impacts of different land uses and land covers on groundwater dynamics, sustainable groundwater resource utilization is promoted, ensuring long-term water security in the study area.
The distance from the main river plays a crucial role in understanding GWPZs within the study area (Fig. 7). As distance from the river increases, various hydrological and geological factors come into play, influencing groundwater dynamics. Near the main river (0–5.0 meters), groundwater potential is typically high. This is because rivers serve as primary sources of groundwater recharge through direct infiltration and lateral movement of water from the river channel into adjacent aquifers. Groundwater levels are often elevated near rivers due to the proximity to surface water bodies, facilitating replenishment. Moving slightly further away (5.1–10.0 meters), the groundwater potential remains relatively high, albeit with some attenuation compared to the immediate river vicinity. Groundwater recharge continues to occur through lateral flow and infiltration processes, albeit at slightly reduced rates. The proximity to the river maintains a strong influence on groundwater levels and quality. In the intermediate zones (10.1–15.0 meters and 15.1–20.0 meters), groundwater potential gradually decreases as the distance from the main river increases. While recharge from the river still contributes to groundwater replenishment, the impact diminishes with distance. Geological factors such as soil permeability and aquifer characteristics become more significant in controlling groundwater dynamics in these zones. Further away from the river (20.1–25.0 meters), groundwater potential reaches lower levels. Recharge from the river becomes negligible, and groundwater dynamics are predominantly influenced by local factors such as rainfall, soil properties, and land use. Aquifer properties and geological formations play a more significant role in determining groundwater availability and quality in these distant zones. Understanding the significance of distance from the main river when studying GWPZs is essential for effective water resource management and sustainable groundwater utilization. By delineating zones based on distance from the river, authorities can prioritize areas with higher potential for groundwater recharge and implement targeted conservation measures to ensure the long-term sustainability of groundwater resources in the study area.
The geomorphological features of the study area, as depicted by variations in height above sea level, provide valuable insights into GWPZs and hydrological processes. Starting with the lowest areas at 920 meters above sea level, which primarily align along the Boteti River, these regions represent floodplains or alluvial plains. Due to their proximity to the river, they are characterized by high GWPZs. The frequent inundation and deposition of sediments contribute to the recharge of aquifers, making these areas favourable for groundwater replenishment. Moving up to areas at 930 meters above sea level, which also border the Boteti River, these zones continue to exhibit high groundwater potential. The gradual rise in elevation from 920 to 930 meters reflects the gentle slope of the floodplain, facilitating lateral movement of groundwater from the river towards adjacent areas. At 940 meters above sea level, the landscape transitions to slightly elevated terraces or floodplain benches. These areas maintain moderate to high groundwater potential, influenced by their proximity to the river and the gradual incline of the terrain. Groundwater recharge continues to occur, albeit at slightly reduced rates compared to the lower floodplain areas. The yellow areas at 950 metres (Fig. 8) above sea level mark higher terraces or elevated plateaus, indicating a further distance from the river channel. While groundwater potential remains moderate in these regions, the recharge processes are primarily influenced by local rainfall and subsurface flow rather than direct river interactions. Moving to the orange-coloured areas at 950 meters and isolated red areas at 960 meters above sea level, represent the highest points within the study area. These elevated zones are typically ridges or upland areas, characterized by lower groundwater potential compared to the lower-lying floodplains and terraces. Groundwater recharge in these regions is primarily limited to localized rainfall infiltration and runoff. Thus, variations in height above sea level within the study area reflect distinct geomorphological features, each with implications for GWPZs. Understanding these features is essential for delineating areas of high, moderate, and low groundwater potential, guiding effective groundwater management and utilization strategies in the study area.
Groundwater Potential Zones Assessment
After determining the final weights for each thematic layer and their corresponding features (Table 1), the layers were transformed into raster format and merged using a raster calculator in ArcGIS 10.8 to delineate the GWPZ. The cumulative scores for the GWPZ (refer to Fig. 9) were then classified into five categories: "very poor," covering 2.8% of the total area; "poor," spanning 44.5% of the investigation region; "moderate," comprising 45.0% of the research area; "good," occupying 7.6% of the study expanse; and "very good," representing only 0.1% of the total study area. Notably, regions along the Boteti River exhibit good groundwater potential (7.6%). Areas near the Boteti River, particularly its floodplains, are characterized by moderate GWPZs (45.0%). The penultimate section of the study area demonstrates poor GWPZs (44.5%), while the largest portion (45%) exhibits moderate GWPZs, with only a minimal fraction (0.1%) designated as very good GWPZs (Table 6)
Table 6
Assignment of Weights and Ranking of Subclasses for the Thematic Layer
Thematic Layer | Assigned Weight | Classes | Rank |
| | 0-60.3 | 5 |
| | 60.3-118.7 | 4 |
Drainage Density | 0.02 | 118.8-178.9 | 3 |
| | 179.0-255.6 | 2 |
| | 255.7-465.5 | 1 |
| | 0-0.18 | 1 |
| | 0.19–0.49 | 2 |
Lineament Density | 0.03 | 0.5–0.8 | 3 |
| | 0.81–1.16 | 4 |
| | 1.17–1.93 | 5 |
| | 0-1.07 | 5 |
| | 1.08–1.87 | 4 |
Slope | 0.32 | 1.9–2.81 | 3 |
| | 2.82–4.26 | 2 |
| | 4.27–12.36 | 1 |
| | 274.9-455.7 | 1 |
| | 455.8-605.2 | 2 |
Precipitation | 0.31 | 605.3-733.8 | 3 |
| | 733.9–911.0 | 4 |
| | 911.5-1161.3 | 5 |
| | eutric fluvisols | 3 |
Soil | 0.04 | cryic soil | 2 |
| | Health shrubs/Trees | 4 |
LULC | | Grasslands | 3 |
| | Built-up Area | 1 |
| 0.06 | Water | 5 |
| | Gravel/Barren/Tarred Road | 2 |
| | 0–5.0 | 5 |
| | 5.1–10.0 | 4 |
Distance from River | 0.16 | 10.1–15.0 | 3 |
| | 15.1–20.0 | 2 |
| | 20.1–25.0 | 1 |
| | 920–930 | 5 |
| | 940–940 | 4 |
Height above sea level | 0.06 | 950 − 940 | 3 |
| | 950–950 | 2 |
| | 960–960 | 1 |